Yamaha YZ250F

[macribs]

This must turn out to be a keeper. Finally someone got their hands on the new 180 from QS. Must about two years since QS first posted about this motor on their site. I will tag along for rest of the build and look forward to see it completed.

Have you decided what controller to use? What kind of peak/burst power are you targeting for this build?

 

[j bjork]

I think this is a keeper too :wink:
I plan to use a sevcon size 6, if I can set it up good:?
Somewhere in the 40kw region would be nice, not that I will be able to use it..

Sometimes I wonder if I should have gone with a 138 70h with the gear reduction.
I want a little more power than what I have on the ktm. But a 100 or so phase amp more, and 30V more might very well have made the difference. It is not often I actually can use more than what the ktm has to give.

With that it would have been a lot easier to fit everything, and the bike could probably have been 10-15kg lighter.
But I went this way, and I think this will be a killer if I can just get it to work properly

 

[j bjork]

About the sevcon, it has found its place on the bike:

With the suspension bottomed out:

There will be cover on the inside of the controller too.
I hope it wont be too tight after all the extras,(wiring and cover) I may very well want a wider tire later.
Now it is a 100/xx/19 on a 1.85" wide rim I think.
I will probably run at least 110/xx/18 on a 2.15" wide rim later, then it might start to get tight..

It is pretty tight, but there are a few mm between controller and battery and brake cylinder.
It also fits behind de side cover after a little trimming :wink:

 

[jhoexp]

You will need a super hard protective cover and it will still be battered by rocks, mud, dirt, etc...

Also very exposed to damage when you tip or fall on the right, like the exhaust was. I don't know you, but I rebuild the exhaus every few seasons, and this is a much more sensitive part.

I understand the need for a big battery, but I think that you really should delete the rear module and put the sevcon under the seat, and probably modify/make a new rear fender to cover it...

 

[j bjork]

You will need a super hard protective cover and it will still be battered by rocks, mud, dirt, etc...

The only problem I see from the inside is if a rock should get stuck between the controller cover and tire.
To be able to damage the controller it would have to bend or brake the cover enough to also hit the controller, and still have enough force to also damage the controller.
Sure it will be possible, but I think it is unlikely to happen.

How big damage do you get on the inside on the muffler?
And what is it made of, 1,5mm alu? 1mm stainless steel?

I think something like 3mm alu or 4 layers of fiberglass would make a more than enough stable cover.

I am more worried about hits from the outside, but I think it would take a pretty hard hit in the lower rear corner to do any damage to the controller. It has to bend my 40x40x4mm frame, and the 20mm thick baseplate on the controller.

It is a big ass controller, it dosent mount easy under the seat even if I sacrifice the rear part of the battery:

 

[Elektrosherpa]

You will need a super hard protective cover and it will still be battered by rocks, mud, dirt, etc...

How big damage do you get on the inside on the muffler?

Well--if a muffler is damaged, it's not a big problem-you produce more noise, and maybe lose a tiny bit of power.
But if the top of the controller with its cables is exposed, or bent against other metalic parts,
quite nasty things can happen-at least in my fantasy

And if you put the controller the way you showed above, your cables will run close to the rear wheel, too.
I wouldnt feel comfortable at all with that setup.

 

[j bjork]

Well, if I make the cover of alu I will put a rubber mat between that and the controller so the metal cant touch the terminals if it gets bent or dented.

What I am curious about is if there is usually big enough hits on the inside of the muffler from rocks or similar to damage it.
Like leave a decent sized bump or so.
I have the muffler from this bike, and it seems to be pretty thin, so I expect my cover to take bigger hits before being seriously damaged.

 

[Elektrosherpa]

What I am curious about is if there is usually big enough hits on the inside of the muffler from rocks or similar to damage it.
Like leave a decent sized bump or so.

On my Bultaco I have already the second or third muffler-and since the engine was a 2stroke one, you can be sure they didnt die of rust...
OK, this is a trials bike, and when I was younger I was doing pretty wild things with it-but when I look at your videos, you dont seem to be a member of the "go slow - go safe"- fraction either... 8)

 

[macribs]

I think this is a keeper too :wink:
I plan to use a sevcon size 6, if I can set it up good:?
Somewhere in the 40kw region would be nice, not that I will be able to use it..

Sometimes I wonder if I should have gone with a 138 70h with the gear reduction.
I want a little more power than what I have on the ktm. But a 100 or so phase amp more, and 30V more might very well have made the difference. It is not often I actually can use more than what the ktm has to give.

With that it would have been a lot easier to fit everything, and the bike could probably have been 10-15kg lighter.
But I went this way, and I think this will be a killer if I can just get it to work properly

You will find a way to fit it all together, you already did welds on the frame. If things don't fit, cut off something and hammer it back into a fitting shape and re weld it back in. 10-15kg lighter? Don't worry too much. If you ain't doing crazy enduro or trial type riding you will get used to it.

Those sevcon are mean bastards to program I hear, but after setup is completed and things runs smooth I am sure you will be pleased.

 

[j bjork]

Yes, I have fitted all the big stuff. Now it is just smaller stuff left, it should not be too much hassle to get the rest in place a hope.

Yesterday I planned to make the mold for the upper part of the battery box, but I ran in to trouble:

First I thought there was a gelcoat problem, but couldn't understand what.
I mixed it well, it wasn't old or exposed to air or something.
After waiting the whole day it had cured enough so I could peel it off:

It wasn't a gelcoat problem, it was a paint problem..
It was only in a few areas, so it is probably that this paint needed a little longer to cure properly.
These areas was probably a little thicker.

I was worried about the inductance in the motor, I read in the axiom thread that another guy had problems running the motor because of low inductance. His vesc based controller had measured about 13uH.
I got my inductance meter today and tried it:

I get about 120-170uH depending on motor position :?
Dosent that seem high instead? Am I doing something wrong?
It seems a zero 75-7 has about 30uH from what I have seen in sevcon DCF files.

 

[hugok]

120-170 uH is about 10 times more than I have on my QS138 if i remember correct.

Have you tried measure the other phases? 120-170 uH doesn't sound like a strange value since the motor is way bigger and I guess its quite low KV.

 

[j bjork]

120-170 uH is about 10 times more than I have on my QS138 if i remember correct.

Have you tried measure the other phases? 120-170 uH doesn't sound like a strange value since the motor is way bigger and I guess its quite low KV.

It dosent seem to make much difference what phases, but I noticed different positions makes a difference.
Before I changed position slightly between "coggs"
But when I change to different between "cogg" position I find a lower value that is around 100uH.

So there are basically a high value of about 150uH, that change between about 120-170 depending if I move the shaft slightly back and forth. Then there is a value of about 100uH, that doesn't change much at all when I move the shaft slightly.

The kv should be about 55, but it I havent confirmed it.

 

[j bjork]

I was curious about these inductance values and did a test on my qs 138 70h on the ktm for comparison.
It is much tighter between "coggs", so there is really no moving the shaft between.
That mean I get more stable values, but I actually get three variants.

About 63uH, 71uH and 85uH. (They can vary a little, maybe about 2 up and down from these values)

It would be interesting to hear some feedback if this seems right, if they are the values on the 180 90h probably are too.
Anyway it seems like the inductance in the 180 90h are about double, good or bad I dont know..

A new try on the battery box:

I didnt want to start over with filler primer and maybe putty, so I only sanded down most of the top layer.
I wiped it with thinner to see if the rest could take it, and it seemed fine.

Now with proper paint. White is not the best as I will use white gelcoat for the mold, so it is harder to see when I have good coverage.
But I didnt want to use up the grey (that is actually for the frame etc. on the ktm) and the other option is flat black.
I think the surface would not be slippery enough on that..

 

[j bjork]

I continued with some more measurements on the motor.

I did some tests of resistance by connecting a load in series with the motor and measure voltage drop.
I tested with two different loads, a little under 8A and a little over 40A.
I got the same results of very close to 5mohm, so I think that is about correct.
I will probably try with a better A meter just in case, but I dont think it will change much.

I then measured back emf, from what I read it is the right way to determine Kv.
I tried some different methods, with a multimeter and with oscilloscope.

From what I read I should take the multimeter value x 1,414 (to convert from rms to peak) then x 0.95.
Then the rpm / with that value.
With the oscilloscope I get this:

This should be at 1332rpm.

The Kv seems to be about 61, I guess if measured with a controller it would be a little less.

 

[hugok]

Have you tried to tool for sevcon? DVT or whats it called.

There you should be able to measure Inductance and resistance.

I use another brand of controllers and they measure this during motor learning.

 

[j bjork]

Have you tried to tool for sevcon? DVT or whats it called.

There you should be able to measure Inductance and resistance.

I use another brand of controllers and they measure this during motor learning.

I have tried it offline, I have not connected the controller.

From what I have seen automatic or learning dosent seem to be a thing on sevcon.
But I probably have missed some functions, and there are surely things I dont see offline.

Made the mold, looks very nice so far:

Not so nice here:

I used one layer gelcoat to save time, but that was a mistake
It usually is enough, but I think a few things played together here.
This gelcoat seems to not cover as good as I am used to. I also have a new hardener, it seems less effective than what I am used to.

I didnt wait enough for the gelcoat to start gelling before laying fiberglass. I am afraid to wait too long, I have had problems in the past with "raising" that I think is because the gelling has gone too far.

 

[JimVonBaden]

Looking great!

 

[Redsea Rider]

Nothing is decided yet, but I am looking at the qs 180 90h.
Qs recommends 750-850 phase amps, and 72-96V.
So I guess I should try to find a controller that can do 1000pA and 96V nominal in that case..

It would be nice with a 6 phase motor, and use dual controllers.

It would be nice if you could go for a ride before summers over

 

[j bjork]

Thank you both
It wont be a testride before the summer is over, that is like 2 weeks away.
But maybe before the winter.
Last year I think we were riding most of November, so there is still hope.
December to late February or early Marsh you need spike tires, and I dont think I will start riding on that with this bike this year.

The top of the battery box fresh out of the mold.
It is a bit rough in the surface, because the mold was..

It is a bit annoying, as I put the work in the model to make it perfect.
But the gelcoat should be pretty thick, so it should be possible to sand and polish to a nice surface.
I will see how it looks on the bike.

 

[tylerwatts]

Very nice J! Your packaging on your builds is well done, just a pity that controller couldn't fit under the seat. I hope it's out the way on the side there.
And the motor looks very promising, congrats.

Sent from my M2102J20SG using Tapatalk

 

[j bjork]

Thank you.
I planned to answer when I had an update, but it seems to take a while..
I am trying to fit everything in the battery now, and it is some work before I have something to show.

It probably wouldn't be impossible to fit the controller under the seat, I just think the sacrifices would be too much.
I would have to sacrifice 6 cells from the battery, and go for 24s instead of 30s.
It would take some room from the wheelhouse(? dont know about this translation) I am not sure how bad that would be, some changes of the rear frame to put the end of the controller almost touching the seat would probably make it work.

I would have to cut off a lot of the rear fender, almost all of the part that is mounted on the frame.
It would take a big reconstitution of the fender, probably making a new one.

Basically not sacrifices I was willing to make. Of course putting everything in the middle of the bike is best, but if you want a big motor, a big battery and a big controller it is not going to happen.
And it is not much use to have one or two of the tree, then you might as well go smaller on all of them.

How ever you build, it is always going to be compromises.
I can only hope it is not coming back to bite me in the end, but you will never know until you know :wink:

 

[j bjork]

It has been a long time since I updated, but the progress has been slow.
Mostly small things on the battery, but I will start with the charger:

Some soldering wires, I drill holes and push the wires trough so I can solder them on both sides.
Both to split the load, and to not risk ripping off the sheet from the pcb.
On the second picture there is also a pot to adjust the voltage.

This one should be finished, I just have to get the program to work so I can set the amps..
That may very well be the hardest part for me :|
Then I have one more to build.
I am modifying after the instructions in this thread:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=66169&hilit=Eaton+APR

I have made some brackets and a series connection.
Then trying to fit everything:

Mostly in the cad state still (cardboard aided design)
Bms, fuse and contactors in place, I need to fit a relay and a shunt too.
And a lot of wiring..

 

[hugok]

What kind of charger are you using?

 

[j bjork]

What kind of charger are you using?

It is server psu:s, eaton apr48-3g.

I am setting them to 60v each, and series connect 2 of them.
But I need to get the program to work to be able to set the amp.

 

[j bjork]

I think it is about time for an update..
There isn't a lot to show, I am just struggling to fit everything in the battery box:

After a lot of trying and adjusting where I thought the problem was, I realized I could just put clay on the interesting places
I wish I did that before I finished welding the brackets, changes would have been easier :lol:

After testing a bunch of shunts:

4 in a line here, I seem to get the same result on them all..

And trying to make a holder:

That I had big problems trying to fit in the battery, I decided to use the shunts from the bms:

Unsoldered from the bms here, cut a line on the pcb to get two sense wires.
One before and one after the shunts. That way it should not matter if I get a little voltage drop in the power wire between the shunt and bms.

And here are the shunts now.

I started a little with bus bars, and decided to make a tab on the battery.
I only have 1mm copper, so I made it double.
One is folded over the tab on the battery, than one on top of the first one.